Global Patent Index - EP 4303768 A1

EP 4303768 A1 20240110 - METHOD AND APPARATUS FOR MULTI-TASK PROCESSING

Title (en)

METHOD AND APPARATUS FOR MULTI-TASK PROCESSING

Title (de)

VERFAHREN UND VORRICHTUNG ZUR MULTITASKVERARBEITUNG

Title (fr)

PROCÉDÉ ET APPAREIL DE TRAITEMENT MULTITÂCHE

Publication

EP 4303768 A1 20240110 (EN)

Application

EP 23167984 A 20230414

Priority

  • KR 20220083717 A 20220707
  • KR 20220110186 A 20220831

Abstract (en)

A method and apparatus for multi-task processing are disclosed. The method includes obtaining a base output corresponding to a first layer, restoring an input map corresponding to a second layer, obtaining an output map corresponding to the second layer, obtaining a delta output map corresponding to the second layer, and storing the base output map and the delta output map.

IPC 8 full level

G06N 3/045 (2023.01); G06N 3/0495 (2023.01); G06N 3/063 (2023.01); G06N 3/082 (2023.01); G06N 3/096 (2023.01); G06V 10/82 (2022.01); G06V 20/10 (2022.01); G06N 3/0464 (2023.01); G06N 3/048 (2023.01); G06N 3/084 (2023.01); G06N 3/09 (2023.01)

CPC (source: EP US)

G06N 3/045 (2023.01 - EP US); G06N 3/0464 (2023.01 - US); G06N 3/0495 (2023.01 - EP); G06N 3/063 (2013.01 - EP); G06N 3/082 (2013.01 - EP); G06N 3/096 (2023.01 - EP); G06V 10/82 (2022.01 - EP); G06V 20/10 (2022.01 - EP); G06N 3/0464 (2023.01 - EP); G06N 3/048 (2023.01 - EP); G06N 3/084 (2013.01 - EP); G06N 3/09 (2023.01 - EP)

Citation (search report)

  • [I] KAWAKAMI REI ET AL: "Cross-Connected Networks for Multi-Task Learning of Detection and Segmentation", 2019 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), IEEE, 22 September 2019 (2019-09-22), pages 3636 - 3640, XP033647593, DOI: 10.1109/ICIP.2019.8803687
  • [A] RHU MINSOO ET AL: "Compressing DMA Engine: Leveraging Activation Sparsity for Training Deep Neural Networks", 2018 IEEE INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE COMPUTER ARCHITECTURE (HPCA), IEEE, 24 February 2018 (2018-02-24), pages 78 - 91, XP033341940, DOI: 10.1109/HPCA.2018.00017
  • [A] DEMI GUO ET AL: "Parameter-Efficient Transfer Learning with Diff Pruning", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 14 December 2020 (2020-12-14), XP081837610
  • [XPI] SHIN JAEKANG JKSHIN94@KAIST AC KR ET AL: "Algorithm/architecture co-design for energy-efficient acceleration of multi-task DNN", PROCEEDINGS OF THE 59TH ACM/IEEE DESIGN AUTOMATION CONFERENCE, ACMPUB27, NEW YORK, NY, USA, 10 July 2022 (2022-07-10), pages 253 - 258, XP058889090, ISBN: 978-1-4503-9739-1, DOI: 10.1145/3489517.3530455

Designated contracting state (EPC)

AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR

Designated extension state (EPC)

BA

Designated validation state (EPC)

KH MA MD TN

DOCDB simple family (publication)

EP 4303768 A1 20240110; US 2024013029 A1 20240111

DOCDB simple family (application)

EP 23167984 A 20230414; US 202318337195 A 20230619